I am writing a bash script, in which I am trying to check if there are particular parameters provided. I've noticed a strange (at least for me) behavior of [ -n arg ]
test. For the following script:
#!/bin/bash
if [ -n $1 ]; then
echo "The 1st argument is of NON ZERO length"
fi
if [ -z $1 ]; then
echo "The 1st argument is of ZERO length"
fi
I am getting results as follows:
with no parameters:
xylodev@ubuntu:~$ ./my-bash-script.sh
The 1st argument is of NON ZERO length
The 1st argument is of ZERO length
with parameters:
xylodev@ubuntu:~$ ./my-bash-script.sh foobar
The 1st argument is of NON ZERO length
I've already found out that enclosing $1
in double quotes gives me the results as expected, but I still wonder why both tests return true when quotes are not used and the script is called with no parameters? It seems that $1
is null then, so [ -n $1 ]
should return false, shouldn't it?
Quote it.
if [ -n "$1" ]; then
Without the quotes, if $1
is empty, you execute [ -n ]
, which is true*, and if $1
is not empty, then it's obviously true.
* If you give [
a single argument (excluding ]
), it is always true. (Incidentally, this is a pitfall that many new users fall into when they expect [ 0 ]
to be false). In this case, the single string is -n
.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With